To test a new hypothesis concerning the mechanism of hypertonic urine formation in the mammalian inner medulla, two sets of partial differential equations will be derived; one that embodies the new theory and another that stimulates the classic countercurrent multiplier model. The models will be implemented to design a pattern of intravenous urea infusion that will maximize the difference in performance between the two theories. The infusion pattern will then be used in an experimental situation to determine which hypothesis is correct. To resolve controversies concerning the control of inner medullary circulation, erythrocyte velocity will be measured in individual capillaries with a two slit method, in different diuretic states. A model of the glomerulus, proximal tubule, peritubular capillary, efferent arteriole, and interstitial space will be developed to evaluate the fluid mechanical peroperties of the system.